A clinical laboratory examines some part of the living patient—his excreta or blood or secretions—to help the physicians reach a diagnosis or provide a better treatment. The oldest example of this, going far back into antiquity, is uroscopy. As the name indicates, in this examination the physician did not manipulate the urine but merely looked at it. Certain appearances allegedly correlated with certain clinical states.
The principle involved—that the characteristics of the urine provided a sign of what was going on within the body—was sound, even though the coefficient of error was enormous. For centuries, physicians studied the urine with various methods, attempting to draw conclusions that would remain valid. Hermann Boerhaave (1668-1738) for example, early in the 18th century, tried to determine whether urine contained "blood serum." He collected a cask of urine whose weight, he said, he reduced by evaporation from 100 lb to 1 lb. But, he